Linux Foundation / CNCF OTCA: OpenTelemetry Certified Associate

Last checked October 22, 2026 against the official Linux Foundation / CNCF exam page.

1. Question 1Observability Concepts and Signals

   Which OpenTelemetry signal records the path of a single request as it moves across multiple services, including timing for each operation?

   1. ALogs only
   2. BProfiles of disk usage
   3. CMetrics
   4. DTraces
   Show answer and explanationHide answer

   Correct answer: D

   Traces capture a request's journey as spans with timing and parent-child relationships. Metrics aggregate numeric measurements, logs record events, and disk profiles are not a core signal.

2. Question 2OpenTelemetry Collector

   In the OpenTelemetry Collector, what is a receiver?

   1. AA component that sends telemetry to a backend over OTLP
   2. BA dashboard panel that renders time series charts from stored metrics for operators
   3. CA component that accepts telemetry data from applications or other sources
   4. DA database index structure, such as a B-tree on a timestamp column, used for lookups
   Show answer and explanationHide answer

   Correct answer: C

   Receivers accept telemetry into the Collector. Exporters send telemetry out, dashboards display data and indexes are database structures.

3. Question 3Instrumentation

   Which instrumentation approach adds telemetry to an application without changing its source code?

   1. AAutomatic instrumentation using an agent or library hooks
   2. BPrinting variable values to the console output with print calls
   3. CManual instrumentation written into every function in the codebase
   4. DEditing the database schema and column types
   Show answer and explanationHide answer

   Correct answer: A

   Automatic instrumentation attaches to libraries or runtimes without code changes. Manual instrumentation requires code edits, console prints are not instrumentation, and schema edits are unrelated.

4. Question 4Sampling and Performance

   A service needs its traces sampled at a fixed ratio, while still propagating the trace context to downstream services so they can make their own decisions. Which behavior is correct?

   1. APropagating the sampling decision in the trace context so downstream services respect it
   2. BIgnoring the sampling decision so each downstream service samples independently of the parent span
   3. CDropping the trace context header at the service boundary before the outbound call
   4. DDeferring sampling to the database layer after spans have been stored
   Show answer and explanationHide answer

   Correct answer: A

   Propagating the sampling decision keeps traces consistent across services. Ignoring decisions creates partial traces, dropping context breaks linkage, and database-only sampling does not apply to the whole trace.

5. Question 5Observability Concepts and Signals

   Which signal is best suited to tracking the total number of HTTP requests per second over time with low storage cost?

   1. AMetrics
   2. BTraces for every request
   3. CUnstructured log lines only
   4. DScreenshots of dashboards
   Show answer and explanationHide answer

   Correct answer: A

   Metrics are aggregated numeric time series that are efficient for rates and counts over time. Tracing every request is expensive for this purpose, and log lines and screenshots are not efficient for rate tracking.